Preparation and characterization of new challenge stocks of SIVmac32H J5 following rapid serial passage of virus in vivo.

BACKGROUND: A new challenge stock of the simian immunodeficiency virus SIVmacJ5 has been produced following passage in vivo. METHODS: SIVmacJ5 3/92 (J5M), was passaged serially through cynomolgus macaques (Macaca fascicularis) by intravenous inoculation of infected spleen cells isolated and prepared 14 days post-infection. Two challenge stocks, SIVmacJ5 S61MLN and SIVmacJ5 S62spl, were prepared by culture of lymphoid tissue ex vivo. RESULTS: These virus stocks appeared better adapted for replication in M. fascicularis as demonstrated by a greater persistence of recoverable live virus from the periphery and increased pathology in lymphoid tissues 20 weeks post-challenge as detected by immunohistochemistry. Sequence analysis of the envelope gene from these stocks did not identify marked diversification of sequence as a result of this procedure. CONCLUSIONS: These stocks display more robust peripheral persistence and tissue pathology in cynomolgus macaques and should prove valuable analysing recombinant vaccines based upon SIVmacJ5 transgenes.

Construction of infectious SIV/HIV-2 chimeras.

OBJECTIVE: To construct SIV/HIV-2 chimeras (SHIV) that replicate in vivo. These would be valuable tools to elucidate the mechanism by which HIV-2 can bypass protection conferred by live attenuated SIV vaccines. METHOD: Novel SHIV were constructed to express either the vpx, vpr, tat, rev and env genes (SHIV-2isy env) or the gag and pol genes (SHIV-2isy gag/pol) of the infectious molecular clone HIV-2isy in an SIVmac backbone. The replication of SHIV-2isy env and SHIV-2isy gag/pol were evaluated on selected cell lines and peripheral blood mononuclear cells (PBMC) in vitro. In addition, their infectivity was assessed in vivo. RESULT: Virus stocks of SHIV-2isy env and SHIV-2isy gag/pol were prepared in vitro. For SHIV-2isy gag/pol both the 5' and 3' boundaries of the chimeric construct were critical for infectivity in vitro. The growth of each chimera on T cell lines in vitro mirrors that of the parental viruses donating the envelope gene. On PBMCs SHIV-2isy env replicated well on human and simian PBMC whereas SHIV-2isy gag/pol replicated to detectable levels on human PBMC only. In vivo, SHIV-2isy env virus was isolated from one of two cynomolgus macaques challenged intravenously, SHIV-2isy gag/pol was isolated from one of two cynomolgus macaques and both rhesus macaques challenged intravenously. CONCLUSION: This is the first report of SIV/HIV-2 chimeras that are infectious in macaques. Moreover, this is the first report of an infectious chimera in which both SIV gag and pol have been replaced with the equivalent regions of an HIV isolate.

The appearance of escape variants in vivo does not account for the failure of recombinant envelope vaccines to protect against simian immunodeficiency virus.

The presence or evolution of immune escape variants has been proposed to account for the failure of recombinant envelope vaccines to protect macaques against challenge with simian immunodeficiency virus (SIVmac). To address this issue, two groups of three cynomolgus macaques were immunized with recombinant SIV Env vaccines using two different vaccine schedules. One group of macaques received four injections of recombinant SIV gp120 in SAF-1 containing threonyl muramyl dipeptide as adjuvant. A second group were primed twice with recombinant vaccinia virus expressing SIV gp160 and then boosted twice with recombinant SIV gp120. Both vaccine schedules elicited neutralizing antibodies to Env. However, on the day of challenge, titres of anti-Env antibodies measured by ELISA were higher in macaques primed with recombinant vaccinia virus. Following intravenous challenge with 10 monkey infectious doses of the SIVmac J5M challenge stock, five of the six immunized macaques and all four naive controls became infected. The virus burdens in PBMC of macaques that were primed with recombinant vaccinia virus were lower than those of naive controls, as determined by virus titration and quantitative DNA PCR. Sequence analysis was performed on SIV env amplified from the blood of immunized and naive infected macaques. No variation of SIV env sequence was observed, even in macaques with a reduced virus load, suggesting that the appearance of immune escape variants does not account for the incomplete protection observed. In addition, this study indicates that the measurement of serum neutralizing antibodies may not provide a useful correlate for protection elicited by recombinant envelope vaccines.

Monoclonal antibodies recognize at least five epitopes on the SIV Nef protein and identify an in vitro-induced mutation.

Eleven monoclonal antibodies (MAbs) to SIV Nef were produced and characterized. Five antibody-binding sites on SIV Nef were identified on the basis of the reactivity of the antibodies with recombinant proteins. Two of the five epitopes were defined using overlapping peptides. A further three epitopes could not be defined with peptides but all antibodies reacted in Western blot, suggesting that the epitopes were at least partially conformation dependent. Antibodies in two of the five epitope groups were further differentiated by competition analysis. The panel of MAbs described is able to distinguish between a number of recombinant Nef proteins currently under investigation in vivo in macaques. Two of the MAbs described are able to distinguish between the Nef protein from pathogenic (J5) and attenuated (C8) strains of SIV, thus providing useful tools for studying the relevance of the Nef protein in the pathogenesis of SIV infection. In FACScan analysis two of the MAbs, KK70 and KK75, were used to identify an in vitro-induced mutation in J5 Nef grown in C8166 cells. Sequence analysis of the phenotypic variants identified a mutation of the tryptophan (TGG) at amino acid 214 to a stop codon (TGA), thus truncating the Nef protein. The functional significance of this observation remains unclear but highlights the need to interpret data with caution if virus has been cultured in vitro even for a short period of time.

Progressive dendritic pathology in cynomolgus macaques infected with simian immunodeficiency virus.

Neuronal pathology in acquired immunodeficiency syndrome (AIDS) is of interest in relation to cognitive impairment in AIDS patients and from the broader perspective of the pathogenesis of neurodegeneration. Cortical dendritic spine loss has been described in patients with AIDS and the aim of this study was to test the hypothesis that similar pathology is present in cynomolgus macaques infected with simian immunodeficiency virus (SIV). These animals develop an AIDS-like illness, but multinucleated giant cell encephalitis is not a feature and CNS virus load is found to be very low. Four animals infected for 2.5-3 months and four infected for 2-3 years were compared with four controls. The Golgi-Cox technique was employed to demonstrate dendritic morphology in the frontal cortex and the diameter of apical dendrites, dendritic spine density and dendritic spine lengths were measured in layer V pyramidal cells. Immunohistochemistry for microtubule-associated protein-2 (MAP-2), MHC class II and glial fibrillary acidic protein (GFAP) was also performed. In infected animals there was progressive spine loss and atrophy of remaining spines with loss of MAP-2 immunoreactivity at late time points. No parallel increase in GFAP immunostaining or MHC-class II expression in microglial cells was seen. We conclude that progressive neuronal dendritic pathology is a feature of SIVmac251 infection of cynomolgus macaques and is apparent relatively early in disease. Furthermore, dendritic abnormalities occur in the absence of either multinucleated giant cell pathology or substantial CNS virus load.

Evaluation of a candidate human immunodeficiency virus type 1 (HIV-1) vaccine in macaques: effect of vaccination with HIV-1 gp120 on subsequent challenge with heterologous simian immunodeficiency virus-HIV-1 chimeric virus.

Human immunodeficiency virus type 1 (HIV-1) envelope vaccines can now be evaluated for efficacy in macaques by challenging with chimeric viruses in which the env, tat and rev genes of simian immunodeficiency virus (SIV) have been replaced by those of HIV-1. Most experiments have so far been conducted using gp120 molecules derived from T-cell-adapted LAI or MN strains of HIV-1, which predominantly use the CXCR-4 co-receptor. These vaccines protect against infection by apathogenic chimeric virus carrying the same envelope sequences. In the experiment described here, four macaques were vaccinated with W61D gp120 derived from a low passage Dutch isolate and capable of inhibiting the binding of MIP1beta to the co-receptor CCR-5. This vaccine was potent, inducing high titres of binding and neutralizing antibodies against the homologous HIV-1 and tenfold lower titres against a heterologous challenge virus (SHIV(SF33)) in which the env, tat and rev genes of SIV had been replaced by those of a San Francisco isolate, HIV-1(SF33). Despite strong immune responses to the vaccine there was no evidence that it protected against challenge with this chimeric virus. The antigenic divergence between vaccine and challenge virus or the increased virulence of the challenge virus may be responsible for the inability of this vaccine to protect against infection by SHIV(SF33).

Changes in neuron size in cynomolgus macaques infected with various immunodeficiency viruses and poliovirus.

Human immunodeficiency virus (HIV) infection leads to clinically significant neuronal pathology, but the underlying mechanism remains unclear. Infection of rhesus macaques with the simian immunodeficiency virus SIVmac251 has been shown to cause atrophy of hippocampal pyramidal cells. The aim of the current investigation was to determine whether SIVmac251 and other viruses with differing abilities to cause immune suppression or encephalitis could cause neuronal atrophy in cynomolgus macaques. Animals infected with SIVmac251 (n = 22), HIV-2 (n = 6). SIVmac239 (n = 7) and poliovirus (n = 10) were investigated, together with 16 controls. Hippocampal pyramidal cell diameter, averaged across the four CA subfields, was reduced by 16.6% in the SIVmac251 group (P < 0.0001) and by 13.3% in the HIV-2 group (P < 0.001), even though the latter virus does not generally cause immunosuppression. Conversely, SIVmac239, which does cause immunosuppression, caused an average neuronal hypertrophy of 6.8% (P = 0.033). Of possible relevance to the different behaviour of the two SIVs is that SIVmac239 is lymphocyte tropic and does not infect CNS microglia in vivo whereas SIVmac251 does. Animals inoculated with poliovirus into the lumbar spinal cord for polio vaccine neurovirulence testing acted as positive controls for CNS inflammation and they also showed an increase in neuronal diameter (4.1%, P = 0.025). The atrophy seen with SIVmac251 and HIV-2 involved all CA subfields but the hypertrophy following SIVmac239 or poliovirus infection was restricted to CA1 and CA2. These observations show a dissociation between the ability of immunodeficiency viruses to cause immune suppression and neuronal pathology and demonstrate that CNS inflammation per se may cause neuronal hypertrophy.

Mechanisms of protection induced by attenuated simian immunodeficiency virus. I. Protection cannot be transferred with immune serum.

To evaluate its role in protection, immune serum was collected from four macaques which were chronically infected with live attenuated simian immunodeficiency virus (SIVmacC8) and had resisted challenge with wild-type SIVmacJ5. The immune serum was transferred to two naive cynomolgus macaques by intraperitoneal injection (11 ml/kg). Four control macaques received an intraperitoneal injection of normal saline. One day later, all macaques were challenged with 10 MID50 of the J5M challenge stock of SIV. After challenge, all macaques became infected as determined by virus co-culture and diagnostic PCR. Virus loads in PBMC at 2 weeks post-challenge were indistinguishable between the two groups of macaques. Thus, the failure of passive immunization to transfer protection indicates that serum components alone are not sufficient to mediate the potent protection obtained using live attenuated vaccines. This is the first time that serum has been transferred from animals known to be protected against superinfection.

The construction and evaluation of SIV/HIV chimeras that express the envelope of European HIV type 1 isolates.

The molecular construction of SIV/HIV-1 chimeric viruses (or SHIVs), provides a means of infecting macaques with immunodeficiency viruses that express the envelope protein of HIV-1. However, to date, most SHIVs produced express the envelope of isolates of HIV-1 that have been passaged repeatedly in T cell lines. We have taken SHIV-4 and replaced an NheI-AvrII fragment that encompasses the gp120 region and the extracellular portion of gp41 with the equivalent region of two European isolates of HIV-1 (ACH320.3.1 and HIV-1Han-2). Neither of these viruses had been passaged in T cell lines for prolonged periods prior to molecular cloning. Virus stocks were prepared of both SHIV constructs. In vitro, the relative ability of each clone to replicate in four T cell lines mirrored closely the pattern observed with the parental virus donating the envelope sequences. In vivo, only one of the chimeric viruses was infectious in cynomolgus macaques and its recovery was transient. The factors that affect the replication of SHIVs in vitro and in vivo are discussed.

Mechanisms of protection induced by attenuated simian immunodeficiency virus. IV. Protection against challenge with virus grown in autologous simian cells.

Attenuated simian immunodeficiency virus (SIV) induces potent protection against infection with wild-type virus, but the mechanism of this immunity remains obscure. Allogeneic antibodies, which arise within animals as a result of SIV infection, might protect against challenge with exogenous SIV grown in allogeneic cells. To test this hypothesis, eight macaques were infected with attenuated SIV and subsequently challenged with wild-type SIV grown in autologous cells or heterologous cells. The results clearly demonstrated that animals infected with attenuated SIV are protected against wild-type SIV grown in autologous or heterologous cells. Thus, the hypothesis that live attenuated SIV protects by the induction of allogeneic antibodies is not tenable.

Strategies for AIDS vaccines.

In the global AIDS epidemic, over half of all infections have occurred in people less than 25 years old resulting in profound social, economic and demographic consequences. Current estimates indicate that the present 15 million HIV infections will increase to over 30 million by the end of the millennium. For most countries a safe and effective vaccine offers the only hope of controlling the spread of this disease. The development of an effective vaccine against HIV is beset with formidable obstacles. Despite these difficulties, substantial progress has been made towards developing effective strategies for vaccination. Human clinical trials and animal models for AIDS, particularly simian immunodeficiency virus (SIV) infection of macaques, have proved invaluable in this quest. Inactivated virus vaccines induced potent protection in this model, but subsequent studies revealed that protection was mediated by antibody to cellular proteins present in the vaccine preparations and on the surface of infecting virions. This surprising observation has provided an alternative and complementary approach to the development of vaccines against HIV in man which is still being pursued. Live attenuated vaccines were initially dismissed as far too hazardous. However, the concept has recently been reexamined in the light of powerful evidence that attenuated SIV induces potent protection against a wide variety of viruses administered by intravenous or mucosal routes and even against challenge with viable virus-infected spleen cells. Efforts are now underway to understand the mechanism of this protection and to attempt to reproduce it by less hazardous means. Considerable effort has been devoted to the development of subunit HIV vaccines, predominantly based on the envelope glycoproteins of the virus. Extensive clinical trials in human volunteers have established that these vaccines are safe and antigenic. However, the immune responses appear to be transient and the antibodies induced do not neutralize the primary isolates of HIV which are circulating in the population. There are now three possible approaches to an AIDS vaccine which are being actively pursued.

Protection by attenuated simian immunodeficiency virus in macaques against challenge with virus-infected cells.

A vaccine against AIDS will probably have to protect against challenge both by viable virus-infected cells and by cell-free virus. Eight cynomolgus macaques infected with attenuated simian immunodeficiency virus (SIV) were challenged (four each) with cell-free and cell-associated SIV. All were protected, whereas eight controls were all infected after challenge. These findings show that live-attenuated vaccine can confer protection against SIV in macaques. Extrapolation to human beings will require extensive evaluation of the safety of attenuated retroviruses. Alternatively, the mechanism of this potent protection must be understood and reproduced by less hazardous means.

PIP: That HIV-1 may be transmitted by virus-infected cells as well as by free-floating virus particles make the development of a vaccine against AIDS particularly challenging. The infection of monkeys (macaques) with simian immunodeficiency syndrome (SIV) is a model for HIV infection in man. The authors report findings from a study in which eight cynomolgus macaques infected with attenuated SIV were challenged with cell-free and cell-associated SIV. All were protected, while eight controls were infected after challenge. The researchers found that live-attenuated vaccine can confer protection against SIV in the monkeys. Applying these findings to vaccine development for humans will, however, require extensive evaluation of the safety of attenuated retroviruses. Otherwise, the mechanism of protection must be understood and reproduced by more safe means.

Fine analysis of humoral antibody response to envelope glycoprotein of SIV in infected and vaccinated macaques.

To characterize the serological response to SIV envelope, induced by vaccination with different envelope immunogens or by SIV infection, plasma samples from 11 cynomolgus macaques infected with simian immunodeficiency virus (SIV) and from 16 macaques vaccinated with three different recombinant envelope proteins were analyzed by (1) ELISA, using a variety of antigens including overlapping peptides encompassing the entire sequence of the envelope protein of SIV, and (2) competition assays, using neutralizing monoclonal antibodies to SIV gp120. Seven regions of SIV envelope were predicted to be antigenic. Peptides representing four of these, in the second and third variable regions (V2 and V3) and the fourth constant (C4) region of gp120 and the Gnann region of gp41, were recognized by the majority of sera from infected and vaccinated animals. Additional antigenic regions were identified in the first and fourth variable domains (V1 and V4) and the carboxy terminus (C5) of gp120 and in three additional regions of gp41. Most infected and vaccinated animals made antibodies that competed with the binding of the three conformational MAbs. Among the vaccinated groups, antibodies induced by vaccination with precursor glycoproteins (gp140 or gp160) recognized several additional gp120 epitopes when compared with antibodies induced by external glycoprotein gp130. Sera from infected animals showed a more restricted gp120 response (17 of 46 peptides recognized) compared to animals vaccinated with precursor glycoproteins (31 peptides recognized). The converse was true for antibodies to gp41. Sera from animals vaccinated with recombinant gp140, produced in insect cells, were the only group that failed to compete with the binding of conformational MAbs. Finally, the development of antibodies to specific epitopes of gp120 and gp41 revealed differences between long-term survivors and nonsurvivors, implying that responses to specific epitopes may be important in conferring resistance to disease progression.

Effects of natural sequence variation on recognition by monoclonal antibodies neutralize simian immunodeficiency virus infectivity.

The determinants of immune recognition by five monoclonal antibodies (KK5, KK9, KK17, Senv7.1, and Senv101.1) that neutralize simian immunodeficiency virus infectivity were analyzed. These five neutralizing monoclonal antibodies were generated to native SIVmac251 envelope glycoprotein expressed by a vaccinia virus recombinant vector. All five recognize conformational or discontinuous epitopes and require native antigen for optimal recognition. These monoclonal antibodies also recognize SIVmac239 gp120, but they do not recognize gp120 of two natural variants of SIVmac239, 1-12 and 8-22, which evolved during the course of persistent infection in vivo (D.P.W. Burns and R.C. Desrosiers, J. Virol. 65:1843-1854, 1991). Recombinant viruses which were constructed by exchanging variable regions between SIVmac239 and variant 1-12 were used to define domains important for recognition. Radioimmunoprecipitation analysis demonstrated that sequence changes in variable regions 4 and 5 (V4/V5) were primarily responsible for the loss of recognition of the 1-12 variant. Site-specific mutants were used to define precise changes that eliminate recognition by these neutralizing antibodies. Changing N-409 to D, deletion of KPKE, and deletion of KEQH in V4 each resulted in loss of recognition by all five monoclonal antibodies. SIVs with these natural sequence changes are still replication competent and viable. Changing A-417 to T or A/N-417/418 to TK in V4 or Q-477 to K in V5 did not alter recognition detectably. These results define specific, naturally occurring sequence changes in V4 of SIVmac that result in loss of recognition by one class of SIVmac neutralizing antibodies.

Where are we now with vaccines against AIDS?

PIP: Extensive studies in experimental animal models and phase I and II clinical trials in humans provide some grounds for optimism about developing a vaccine for AIDS. Over 200 monkeys have been protected by simple inactivated vaccines against infection with a lethal challenge dose of simian immunodeficiency virus (SIV). Passive transfer of the antibody to SIV has been shown to prevent infection. Recent vaccination with attenuated, live SIV protected against a challenge with 1000 monkey infectious doses of virus. Studies in chimpanzees have been limited to the unrepresentative IIIB strain of HIV-1. Purified recombinant envelope protein either as the gp 120 surface unit or as the entire gp 160 protein has protected a proportion of chimpanzees against infection. Several experimental immunogens have been tested in phase I and II clinical trials in human volunteers. 5 different recombinant HIV envelope subunits, live recombinant vaccinia virus expressing HIV envelope, and synthetic peptides or virus-like particles derived from yeast, but expressing HIV core proteins, have been used for vaccines that have proved safe. However, large quantities of the subunit HIV antigens are required, and the humoral immune responses are short-lived. Trials with the live recombinant vaccinia virus expressing HIV envelope represent the first use of live recombinant virus in humans. Further studies are under way in France with an avian poxvirus vector, which is host restricted and therefore potentially safer. STudies with combinations of live recombinant virus vaccines followed by immunization with purified subunits indicate that this approach may stimulate both cellular immunity and neutralizing antibodies at higher concentrations than previously observed. The WHO has identified sites in Brazil, Rwanda, Thailand, and Uganda for large scale trials of the efficacy of AIDS vaccines that will probably begin within 5 years.^ieng

Neutralizing antibodies modulate replication of simian immunodeficiency virus SIVmac in primary macaque macrophages.

Cultured macaque macrophages are permissive for the replication of SIVmac251, and inoculation with virus is followed by the production of viral p27. Neutralizing macaque polyclonal and murine monoclonal antibodies preincubated with the virus prevented infection but did not prevent cytopathic virus replication when added more than 3 days after inoculation with virus. However, application of the neutralizing antibodies to macrophages 24 h after inoculation with virus resulted in sustained, low-level production of viral antigen. Cell lysates and individual macrophages from treated cultures contained less viral protein by Western blot (immunoblot) and immunocytochemistry than untreated controls. In situ hybridization and polymerase chain reaction procedures for detecting and estimating relative amounts of viral RNA and DNA showed that both viral nucleic acids failed to increase beyond the levels obtained before the addition of neutralizing antibodies. The data suggest that macrophages may need to be infected with a minimum threshold of virus particles in order to reach their full potential for virus replication and that their exposure to neutralizing antibodies prior to reaching this threshold resulted in limited virus replication.

Protective epitopes on the fusion protein of respiratory syncytial virus recognized by murine and bovine monoclonal antibodies.

The regions of the fusion protein of respiratory syncytial virus (RSV) that react with neutralizing, fusion-inhibiting and highly protective bovine and murine monoclonal antibodies (MAbs) were mapped by two methods: (i) competitive binding assays and (ii) production and analysis of antibody-escape mutants. Competitive binding assays with 16 murine and 10 bovine MAbs identified 11 antigenic sites on the fusion (F) protein, many of which overlapped extensively, and indicated that cattle, a natural host for RSV, and mice recognize similar epitopes. Neutralizing MAbs identified four sites, two of which were also fusion-inhibiting and highly protective in mice. The pattern of reactivity of antibody-escape mutants with the MAbs confirmed the mapping of the protective epitopes deduced from competitive binding assays. A comparison of the biological properties of MAbs to the F protein indicated that protection against RSV infection correlated with fusion inhibition rather than neutralization titre or complement-dependent lysis of virus-infected cells.